BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//https://techplay.jp//JP
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALDESC:High-dimensional Statistical Modeling Team Seminar (Talk by Pr
 of. Sébastien Fries\, Vrije Universiteit Amsterdam)
X-WR-CALNAME:High-dimensional Statistical Modeling Team Seminar (Talk by Pr
 of. Sébastien Fries\, Vrije Universiteit Amsterdam)
X-WR-TIMEZONE:Asia/Tokyo
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
BEGIN:STANDARD
DTSTART:19700101T000000
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:814135@techplay.jp
SUMMARY:High-dimensional Statistical Modeling Team Seminar (Talk by Prof. S
 ébastien Fries\, Vrije Universiteit Amsterdam)
DTSTART;TZID=Asia/Tokyo:20210511T160000
DTEND;TZID=Asia/Tokyo:20210511T170000
DTSTAMP:20260507T101628Z
CREATED:20210407T140010Z
DESCRIPTION:イベント詳細はこちら\nhttps://techplay.jp/event/81413
 5?utm_medium=referral&utm_source=ics&utm_campaign=ics\n\nTitle:  Path pre
 diction of aggregated α-stable moving averages using semi-norm represent
 ations\n\nAbstract:\nFor (Xt) a two-sided α-stable moving average\, this
  paper studies the conditional distribution of future paths given a piece
  of observed trajectory when the process is far from its central values. 
 Under this framework\, vectors of the form Xt = (Xt−m\,...\,Xt\,Xt+1\,.
 ..\,Xt+h)\, m≥0\, h≥1\, are multivariate α-stable and the dependence
  between the past and future components is encoded in their spectral meas
 ures. A new representation of stable random vectors on unit cylinders –
 sets {s∈Rm+h+1: ‖s‖= 1} for ‖·‖ an adequate semi-norm– is pr
 oposed in order to describe the tail behaviour of vectors Xt when only th
 e first m+ 1 components are assumed to be observed and large in norm. Not
  all stable vectors admit such a representation and (Xt) will have to be 
 «anticipative enough» for Xt to admit one. The conditional distribution
  of future paths can then be explicitly derived using the regularly varyi
 ng tails property of stable vectors and has a natural interpretation in t
 erms of pattern identification. The approach extends to processes resulti
 ng from the linear combination of stable moving averages which feature mu
 ch richer dynamics and applied to several examples.\n\nBio: \nSébastien 
 Fries is a tenure-track Assistant Professor at the Department of Economet
 rics and Data Science of the Vrije Universiteit Amsterdam\, and a Marie S
 kłodowska-Curie Fellow. He holds an Engineer’s degree from ENSAE Paris
 \, a M.Sc. in Economics from Paris School of Economics and he completed a
  PhD in Mathematics at Paris-Saclay University in 2018 focusing on the pr
 obabilistic prediction theory of so-called anticipative\, or noncausal\, 
 heavy-tailed processes which remains his main research line. He teaches a
 pplied and theoretical machine learning at bachelor and master level.
LOCATION:オンライン
URL:https://techplay.jp/event/814135?utm_medium=referral&utm_source=ics&utm
 _campaign=ics
END:VEVENT
END:VCALENDAR